With a computer system, it is important to constantly backup data so that even when a hard disk device or other data storage device happens to malfunction, the data can be salvaged. Though data backup can be performed as a manual task by an operator, there is the danger that if an unforeseen accident occurs when the operator has forgotten to perform the backup task, the recovery of data is impossible. Thus since priorly, arrangements, by which data backup is performed automatically by any of various methods without the operator recognizing whatsoever that backup is being performed, have been incorporated in computer systems.
For example, in an operation mode called “level 1” in a disk array system generally referred to as a RAID (Redundant Array of Inexpensive Disks), a disk mirroring process is performed. That is, two hard disk devices of the same capacity are prepared and a process, in which data to be stored are always stored in both of the hard disk devices, is executed. Thus even if a problem occurs with one of the hard disk devices, the recovery of data is enabled by the other hard disk device and the redundancy of data storage can thus be increased.
Methods, in which a normal data storage unit (original file area) and a backup data storage unit (copy file area) are prepared in advance, data are normally stored into the original file area, and the data in the original file area are copied into the copy file area at predetermined timings to increase the redundancy, are also used. For example, Japanese Unexamined Patent Publication No. 2003-263354 discloses a scheduling method for performing automatic backup periodically.
The abovementioned conventional automatic backup methods are intended to increase the redundancy of an entire storage region by copying the entirety of the storage region, which has been designated in advance, as it is. For example, in the disk mirroring mode (level 1) of a RAID system, the contents of one hard disk device are copied exactly as they are into the other hard disk device. In the automatic backup system, in which the contents of the original file area are copied into the copy file area at predetermined timings, the entireties of specific hard disk devices or specific folders are designated as the original file area and the copy file area.
The conventional automatic backup methods are thus premised on designating a hard disk device or a folder as a backup object, and the entire contents of the hard disk device or folder are copied as they are. Data that an operator primarily does not consider necessary to back up thus also become subject to the automatic backup. An unnecessary load is thus placed on the computer system and an efficient backup process cannot be performed.
Recently, due to OS programs and application programs being made highly functional and large in size, various data files are prepared inside a computer system and stored in a hard disk device, etc. Among these various data files, there is quite an amount of files, called “temporary files,” that are prepared just for temporary processes. With the conventional automatic backup method, since a backup setting can only be made in file area units that include even such data that do not need to be backed up, an unnecessary load is placed on the computer system and inefficient operation is made inevitable.
An object of the present invention is to provide an automatic data backup method, with which settings are made in a more finely tuned manner to lighten the load of a system and enable an efficient automatic backup process to be performed.